Expression of Keratin 12 and Maturation of Corneal Epithelium during Development and Postnatal Growth

Abstract

purpose. To determine the kinetics of corneal epithelial maturation during embryonic development and postnatal growth.

methods. Expression patterns of keratin (K)12 and K14 were determined in mouse embryos (embryonic days [E]15.5–19.5), corneas of postnatal day (P)0 to 10 months, and healing corneas after epithelial debridement in P30 and P90 mice. The expression of alkaline phosphatase (AP) was determined during postnatal growth and healing of epithelial debridement of Krt12 ^Cre/Cre /ZAP bitransgenic mice.

results. During embryonic development, K12 expression by corneal peridermal epithelium commenced at E15.5. In the period from E15.5 to P10, the expression of K12 was restricted to the suprabasal and/or superficial cells of the corneal epithelium, whereas the K14 expression was restricted to the basal cells. After P30, K12 expression was sporadically detected in the basal corneal epithelium, and the number of K12-positive basal cells increased as the mice grew older. The number of K14-positive cells that coexpressed K12 increased with age and reached a plateau after P180. Healing of the debrided epithelium facilitated the increase in K14-positive cells that coexpressed K12. Many basal cells of Krt12 ^Cre/Cre /ZAP mice remained undifferentiated and expressed LacZ at P15, and they then differentiated to express Cre, which leads to excision of LacZ and AP expression.

conclusions. In the mouse, the corneal epithelium does not become fully mature until 3 to 6 months after birth, in that a significant number of corneal basal epithelial cells of young mice (<P30), which derive from embryonic surface ectoderm remain undifferentiated and serve as corneal epithelial progenitor cells. These progenitor cells may have some stem cell characteristics.

The corneal epithelium is a stratified squamous epithelium that differs from other stratified epithelia (e.g., the epidermal epithelium), in that it does not form a cornified envelope. Cornea-type epithelial differentiation is signified by the expression of the K12/K3 keratin pair of intermediate filaments, which are essential for corneal epithelial integrity. ^{1 2 3 4} Like other stratified epithelia, the superficial corneal epithelial cells slough off as a result of terminal differentiation. Thus, the corneal epithelial cell population must be replenished by the division of stem cells. In adults, the basal layer of limbus contains slow-cycling and K12-negative cells, many of which are p63 positive, characteristic of stem cells. ^{5 6 7} Therefore, based on the unique geographic location of stem cells, the centripetal movement of corneal epithelial cells has been the central dogma for describing how the corneal epithelium maintains its homoeostasis through life—the so-called X, Y, Z hypothesis. ^{8 9} However, it has recently been suggested that the mouse corneal epithelium does not become mature until 3 months after birth, a plausible hypothesis that is consistent with the mosaic expression patterns of green fluorescent protein (GFP) in the corneal epithelium of young (<3 months) and the spiral enhanced (E)GFP pattern in corneas of adult (>3 months) X-linked CAG-EGFP transgenic mice. ¹⁰ The formation of the spiral EGFP pattern coincides with the time when corneal stem cells become restricted to the basal limbal layer. Thus, it is very likely that during eye growth in young mice, corneal epithelial progenitor cells (stem cells and/or stem-like cells) may exist in the basal layer of the corneal epithelium in addition to the limbal basal layer. However, there is no direct evidence to support this proposition. 

During embryonic development, the differentiation of ocular surface ectoderm gives rise to epithelia of lens, cornea, conjunctiva, eyelid epidermis, lacrimal gland, and Meibomian gland. ¹¹ A two-cell–layered corneal peridermal epithelium forms when the lens detaches from the surface ectoderm during embryonic development at embryonic days (E)11.5 to 12.5. This peridermal epithelium does not express K12. It has been demonstrated that at E15.5, K12 expression is detected by immunofluorescent staining and in situ hybridization. ^{12 13} We previously demonstrated that K12 expression was concomitant with the commencement of corneal epithelium stratification at postnatal day (P)4 to a five- to six-cell–layered epithelium at P14 when the eye opens. ³ In adult mice, the K12 is expressed by all corneal epithelial cells and limbal suprabasal epithelial cells, except the limbal basal cells. ² The terminal differentiation of corneal epithelium involves the upward migration of basal cells and the formation of tight junctions between the superficial cells (occludin positive). ^{14 15} It remains unknown whether all basal cells of the corneal peridermal epithelium derived from surface ectoderm simultaneously differentiate at birth or whether they gradually become differentiated through a period of postnatal life and become characterized by K12 expression that signifies cornea-type epithelial differentiation. 

In the present study, we took advantage of the coexpression of K12 and K14 by basal corneal epithelial cells as a marker signifying the commitment of terminal corneal-type epithelium differentiation, whereas expression of K14 alone is characteristic of progenitor cells of corneal epithelium during embryonic development and postnatal growth of the mouse eye. ^{12 16} Our results indicate that many basal corneal epithelial cells derived from corneal peridermal epithelium of surface ectoderm do not terminally differentiate until 3 to 6 months after birth. 

Materials and Methods

Animals

Animal care and use conformed to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The Institutional Animal Care and Use Committee (IACUC) of the University of Cincinnati approved all animal protocols. 

Cell Suspension

Mice of different ages were killed by CO₂ asphyxia, and the dissected corneas were incubated with 0.5% dispase II in phosphate-buffered saline (PBS) for 30 minutes at 37°C. Epithelium was removed with forceps and further incubated at 37°C for 15 minutes in PBS containing 0.2% trypsin and 1 mM EDTA. The cells were dispersed with a 25-gauge needle, washed with 10% fetal calf serum (FCS) in PBS twice by centrifugation at 1000 rpm for 3 minutes, and suspended in 10 μL FCS. A smear was prepared with 1 μL of the cell suspension on each slide and was air dried at room temperature. The number of cells expressing K12 and K14 was determined by indirect immunofluorescent staining with rabbit anti-K12 ¹⁷ and mouse monoclonal anti-K14 (Sigma-Aldrich, St. Louis, MO) antibodies followed by secondary antibody FITC- and Alexa555-conjugates (Molecular Probes, Eugene, OR), respectively. Micrographs were taken from five random chosen fields of each slide with a fluorescence microscope (model E800; Nikon, Tokyo, Japan). The data were analyzed on computer with Student’s t test (StatView software; SAS Institute Inc., Cary, NC) and are presented as the mean ± SEM. 

Immunohistochemistry

Embryos, obtained by caesarean section, and postnatal eyes were fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (PB; pH 7.4) at 4°C overnight and paraffin embedded as previously described. ¹⁸ Five-micrometer sections were then mounted (Super Frost slides; Fisher Scientific, Pittsburgh, PA). The sections were deparaffinized and hydrated in a graded ethanol series (90%, 80%, and 70% ethanol and PBS for 10 minutes each). The sections were then soaked in PBS for 5 minutes and incubated with blocking buffer containing 4% dry milk in PBS for 1 hour at room temperature and incubated overnight with primary anti-K12N (N-terminal peptide) and anti-K12C (C-terminal peptide) ¹⁷ and anti-K14 antibodies (Sigma-Aldrich) in a moist chamber at 4°C. After washes with PBS, the sections were incubated with the appropriate Alexa-labeled secondary antibodies (Molecular Probes) at room temperature for 45 minutes in a moist chamber, washed three times with PBS for 15 minutes each, and mounted in mounting medium. The images were then captured with a confocal microscope (model LSM510; Carl Zeiss Meditec GmbH, Oberkochen, Germany). 

Wound Healing

Experimental animals were anesthetized by intraperitoneal administration of ketamine hydrochloride (2 mg/g body weight) and xylazine (0.4 mg/g body weight). One drop of topical proparacaine (Alcaine; Alcon, Fort Worth, TX) was applied to each eye before surgery. A central circular 2-mm corneal epithelial debridement (demarcated by a trephine) was created with a corneal rust ring remover with a 0.5-mm burr (Algerbrush II; Alger Equipment Co., Inc., Lago Vista, TX) under a SV11 stereomicroscope (Carl Zeiss Meditec GmbH). ¹⁹ The animals were killed at different intervals after epithelial debridement, and an epithelial cell suspension was prepared as described earlier. 

Wholemount X-gal and AP Staining

Mouse eyeballs were fixed in 0.2% glutaraldehyde, 50 mM EGTA (pH 7.3), and 100 mM MgCl₂ in PBS (pH 7.4) at 4°C overnight. Samples were washed three times for 15 minutes each in LacZ wash buffer (2 mM MgCl₂, 0.01% sodium deoxycholate, and 0.02% Nonidet-P40 [NP-40] in PBS [pH 7.4]). Staining was performed in 0.5 mg/mL X-gal, 4.5 mM potassium ferrocyanide, and 4.5 mM potassium ferricyanide in LacZ wash buffer at 25°C for 12 hours. ²⁰ For alkaline phosphatase staining, specimens were rinsed in PBS, and endogenous alkaline phosphatases were heat inactivated by incubation at 70°C for 30 minutes. ²¹ Alkaline phosphatase activity was visualized with a Fast red kit (Sigma-Aldrich) according to the manufacturer’s protocol. 

Preparation of K12 ^Cre/Cre /ZAP Bitransgenic Mice

The K12 ^{Cre /Cre} mice, which express Cre recombinase in the corneal epithelium, were created by inserting an IRES-Cre minigene into the site immediately behind the stop codon in exon 8 of the Krt12 allele by a knock-in strategy of gene-targeting techniques (Hayashi Y, manuscript in preparation) similar to that for the creation of K12 ^rtTA/rtTA knock-in mice. ¹³ The modified Krt12 ^Cre allele encodes a bicistronic mRNA for the synthesis of K12 and Cre recombinase; thus, the expression of Cre is under the identical regulatory mechanism of K12 expression and does not perturb the synthesis of K12 by corneal epithelium during embryonic development and after birth, similar to what has been shown in K12 ^{rtTA /rtTA} knock-in mice. ¹³ The K12 ^{Cre /Cre} mice were crossbred with ZAP mice ²¹ to obtain K12 ^{Cre /Cre} /ZAP bitransgenic mice. 

Results

During embryonic development, K12 expression by the suprabasal corneal epithelial cells commenced at E15.5, peaked at E17.5 (Fig. 1)and then declined at E19.5 (data not shown), as detected by anti-K12N and anti-K12C antibodies (both antibodies showed identical fluorescence patterns). ¹⁷ After birth, K12 expression remained very low at P0, P2 and then gradually increased at P4, a time point when corneal epithelium begins to stratify. ³ A few K14-positive cells were detected at E15.5 in the basal corneal epithelial layer, their number increased at later developmental stages, and the expression level of K14 was maintained constant in the P0 to P2 stage. At P4, all corneal basal cells were K14 positive but K12 negative. This keratin expression pattern during development and early postnatal growth is similar to that in the limbus in adult mice. This observation indicated that these basal corneal cells did not commit to terminal differentiation at this age and may serve as progenitor cells for the formation of stratified corneal epithelium (Fig. 1) . A significant number of basal corneal epithelial cells originating from surface ectoderm remained K12 negative and presumably undifferentiated at P30 and P90 after birth (Fig. 2) . 

To further determine the kinetics of corneal epithelium differentiation, we counted the number of cells expressing K12 and K14, by double-immunofluorescence staining with epithelial cell suspensions prepared from corneal epithelium of mice at various postnatal ages. Figure 3Ashows representative double immunostaining of corneal epithelial cells prepared from P15 and P300; K14-positive cells were red and K12-positive cells were green. The cells that coexpressed K14 and K12 appeared white. The percentage of K14 cells that also expressed K12 was determined by dividing the number of white cells (K12 and K14) by the number of red cells (K14) by using image-analysis software (ImagePro; Media Cybernetics, Silver Spring, MD). Figure 3Bdemonstrates the percentage of K14-positive cells that coexpressed K12 increased as a function of age, which reached a plateau of 70% at P180 (6 months after birth). The remaining 30% of K14-positive cells represented limbal basal cells and possibly conjunctival basal epithelial cells (K4 positive) that also expressed K14 but not K12. ¹² Immunostaining with anti-K4 revealed that the number of K14-positive cells that coexpressed K4 was small (4.4%). The results indicate that many corneal basal cells derived from surface ectoderm do not commit to terminal differentiation and maintain progenitor cell characteristics until 6 months after birth. 

To verify that many basal cells remain undifferentiated for a relatively long period after birth, epithelial debridement was applied to remove the central corneal epithelium so that the regenerated epithelium would consist of basal cells committed to terminal differentiation. ¹⁷ Figure 4Ashows that the number of cells committed to terminal differentiation increased as determined by coexpression of K12 and K14 after epithelial debridement in P90 mice. The percentage of K12/K14-positive cells of all K14 positive cells was 50.1% ± 7.27% at 0 hours, 55.1% ± 10.3% after 24 hours, 56.3% ± 7.6% after 48 hours, 59.8% ± 7.1% after72 hours, 68.5% ± 7.5% after 4 days, 70.8% ± 7.2% after 5 days, and 70.8% ± 10.2% after 7 days. Epithelial debridement caused an increase in the number of cells that committed to terminal differentiation from 50% to 70% within 4 to 7 days of injury, the same as that of mature corneas at P180. Similar results were obtained in experiments using P30 mice (Fig. 4B) . The percentage of K12/K14-positive cells of all K14-positive cells was 34.6% ±1.5% after 24 hours of wound healing, and 69.7% ±4.3% 7 days after injury. The contralateral uninjured corneas contained 36.4% ± 1.3% and 37.5% ± 12.5% of cells committed to terminal differentiation, in 24 hours and 7 days, respectively, after debridement of the fellow experimental eyes. 

To examine further the time course of corneal epithelium maturation, we took advantage of Krt12 ^{Cre /Cre} /ZAP bitransgenic mice. The Krt12 ^{Cre /Cre} /ZAP mice were obtained by breeding K12-Cre and ZAP mice. We prepared K12-Cre knock-in mice via gene-targeting techniques. The ZAP mice carry LacZ gene flanked by loxP elements followed by an alkaline phosphatase reporter gene, and they normally express LacZ driven by a chicken β-actin promoter. On the excision of LacZ gene by Cre recombinase, the AP is then expressed. ²¹ Thereby, those cells in corneal epithelium that do not express K12 will express the LacZ gene and are positive by X-gal staining for β-galactosidase, whereas those that have differentiated to express K12 will also express Cre recombinase and lead to the excision of LacZ followed by the expression of AP. Figure 5Ashows a mosaic expression pattern of LacZ (blue) and AP (red) in the corneal epithelium of Krt12 ^{Cre /Cre} /ZAP mouse at P15. At P30, the tissues still contained more blue-stained than red-stained cells (Fig. 5B) . At P60, K12- positive cells (red) were in the central cornea, whereas blue cells were in the periphery of the cornea (Fig. 5C) . At P90 and P180, almost all cells in the cornea were red except in the limbus, indicating corneal epithelium maturation, but some blue-stained cells were still present (Figs. 5D 5E) . At P300 no blue cells were visible in the cornea (Fig. 5F) , except in the limbus. As the mice grew older, the number of LacZ-expressing cells decreased, with most LacZ-positive cells located at the periphery and the limbus. This observation implies that the central corneal basal cells commit to terminal differentiation earlier than those at the periphery. The LacZ-positive cells became restricted to the limbus in corneas of mice older than 90 days, consistent with the fact that corneal epithelial stem cells reside in the limbus of adults. 

Epithelium debridement was performed with P30 Krt12 ^Cre/Cre /ZAP mice in an attempt to verify that the removal of the central epithelium would accelerate epithelial maturation. Twenty-four hours after epithelial debridement, the LacZ expression showed a mosaic pattern with a greatly reduced number of LacZ-positive cells in the periphery (compare Fig. 6Awith 6B). After 7 days, a LacZ-expressing pattern of blue stripes was observed, with a further reduction of LacZ-positive cells in both the periphery and central cornea (compare Fig. 6Cwith 6D). 

Discussion

During embryonic development, the corneal peridermal epithelium consists of two cell layers: basal and suprabasal. In the present studies, we used immunofluorescence staining with anti-K12N and K12C antibodies and demonstrated that K12 expression by suprabasal corneal peridermal epithelium commenced at E15.5, peaked at E17.5, and declined afterward during embryonic development, and remained at a low level after birth, at P0 and P2 (Fig. 1) . The observations are consistent with our previous studies showing the expression of K12 by in situ hybridization. ¹³ K12 expression then increases at P4, at a time when the corneal epithelium begins stratification, as shown previously by Liu et al. ² It is notable that, during this embryonic period and in early postnatal life (P0–P4), K12 expression was absent from the basal layer of the corneal epithelium until P10. In contrast, K14 expression maintained a relatively constant level in all corneal and limbal basal cells beginning at E14.5 and thereafter. It has been shown that K14 is also expressed by the progenitor cells residing in the basal layer of most stratified epithelia (e.g., epidermis and conjunctiva). Its expression is mutually exclusive from that of keratins such as K10 and K13 by suprabasal and superficial cells of terminal differentiated epithelia (i.e., epidermis and conjunctiva epithelium, respectively). ¹⁶ In cornea, the coexpression of K14 and K12 by basal epithelial cells signifies the commitment of terminal cornea-type epithelium differentiation. Thus, the lack of K12 expression suggests that the basal epithelial cells maintain progenitor cell characteristics and do not commit terminal corneal-type epithelium differentiation during embryonic development and early postnatal life before P10. The percentage of corneal basal cells that coexpressed K14 and K12 gradually increased with age after P10 and did not reach a plateau until between P90 and P180. This observation implies that many cells originating from the surface ectoderm remain undifferentiated in the basal corneal epithelium of young mice for a relatively long period after birth. Thus, the percentage of corneal basal cells coexpressing K14 and K12 can be used as an index of corneal epithelium maturation. It has been suggested that corneal epithelium becomes mature after P90. ¹⁰ This suggestion is further supported by the results shown in Figures 3 and 5 . The finding that the LacZ-AP staining pattern of Krt12 ^{Cre /Cre} /ZAP mouse changed from mosaic to a full AP-positive cornea also demonstrates corneal epithelial maturation. However, even in the central part of the mature corneal epithelium, a few cells did not differentiate to express K12 and remain LacZ positive (Fig. 5E) . Removal of central corneal epithelium accelerated its maturation, an observation consistent with the notion that epithelial cells of ectodermal origin are quickly replaced by cells from limbus (Figs. 4 6) . It remains unknown whether these K12-negative basal epithelial cells possess self-renewing stem cell characteristics. More studies are needed to examine this possibility further. Our observation that it takes a relatively long time for corneal epithelium to become fully mature after P90 provides an explanation of the higher efficiency of age- and geographically (peripheral versus central) related clonal growth of corneal epithelial cells. ^{9 22 23 24}  

P63 has been suggested to be a marker of limbal stem cells. ⁵ However, the expression of p63 in limbal basal cells has been controversial, as many studies reported p63 expression by basal corneal cells. Because p63 has multiple isoforms, detection of p63 in the limbal basal cells was attributed to the presence of a specific p63 isoform in this region. ^{6 7 25} The presence of progenitor cells in the corneal epithelium of young individuals may provide an alternative explanation for the detection of p63-positive cells in the basal layer of the corneal epithelium. 

The molecular and cellular mechanisms that account for our observation in which the differentiation of central basal corneal epithelial cells precedes those at the periphery remain elusive. It is known, however, that the tissue environment of the so-called stem cell niches has a pivotal role in the maintenance of stem cells in adults. ²⁶ One common feature of this niche tissue environments is that these tissues are usually highly vascularized and consist of unique extracellular matrix components in the basement membrane. ^{27 28} It is likely that the nutrients and various cytokine and growth factors from blood circulation are necessary for the maintenance of the stem cell population in tissues. ⁷ For example, hemopoietic stem cells are in the sinusoid of bone marrow, spleen, and liver, which is highly vascularized. ^{29 30} The corneal stroma is avascular, whereas the limbal stroma is vascularized. Thus, the supplies of nutrient, cytokine, and growth factors to cells at central cornea via circulation are limited in comparison to those at the periphery of cornea and limbus. Of note, it has been demonstrated that the distribution of type IV collagen isoforms in basement membrane underlying cornea epithelium differs from that in the limbus. ^{31 32 33} It is plausible to hypothesize that the vasculature and unique composition of the basement membrane of the limbus may constitute a stem cell niche environment. However, it remains unknown how the vasculature and basement membranes of cornea and limbus are formed differentially during embryonic development. 

 Supported in part by National Eye Institute Grants EY14207 and EY10556; Research to Prevent Blindness (RPB), Inc.; and the Ohio Lions Eye Research Foundation. WWYK was a recipient of an RPB Senior Scientific Investigator Award.

 Submitted for publication September 3, 2005; revised October 24, 2005; accepted December 14, 2005.

 Disclosure: N. Tanifuji-Terai, None; K. Terai, None; Y. Hayashi, None; T. Chikama, None; W.W.-Y. Kao, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Immunofluorescent staining of K14 and K12 in mouse corneas. Eyes dissected from mouse embryos and postnatal mice were fixed in 4% paraformaldehyde and paraffin embedded. Sections (5–7 μm) were subjected to indirect immunofluorescent staining with anti-K12N (and anti-K12C; data not shown). ¹⁷ and K14 antibodies, followed by secondary FITC-conjugates (green) and counterstained with propidium iodide (red). K12 (green) expression by suprabasal epithelial cells commenced at E15.5. At P0, K12 expression was reduced and hardly detected. After P4, K12 expression was gradually upregulated and primarily limited to suprabasal and superficial layers. K14 expression was detected at E15.5 in the basal cells and maintained a constant level throughout the study period (E15.5–P10).

Figure 2.

 

Immunofluorescent staining of K12 in P30 and P90 mouse corneas. Paraffin-embedded sections of P30 and P90 mouse eyes were subjected to indirect immunofluorescent staining with anti-K12, as described in Figure 1 . ( ) K12-negative basal corneal epithelial cells.

Figure 3.

 

Double immunostaining of a corneal epithelial cell suspension prepared from mouse postnatal corneas. Individual epithelial cells in 10 μL of FCS were prepared from corneas of five mice of each age group by dispase II and trypsin treatment. One microliter of cell suspension was spread on a slide and subjected to double immunostaining of K12 (green) and K14 (red). Ten slides were prepared from each group. Five randomly chosen fields were analyzed from each slide. Cells coexpressing K12 and K14 appeared white when the images were merged. The percentage of K14-positive cells that also expressed K12 was determined by dividing the number of white cells by the number of red-stained ones. The number of K12- and K14-positive cells increased and reached a plateau (70%) at P180. (A) Representative immunostaining of P15 and P300 corneal epithelium. (B) The percentage of K12- and K14-positive cells in the total K14-positive cell population of mice at different ages.

Figure 4.

 

Corneal epithelium debridement facilitated epithelium maturation. Corneal epithelium debridement (2 mm in diameter) was performed in one eye of P90 and P30 mice; the other eyes served as uninjured controls. Five injured and five control corneas were used for each experimental group. The number of K12/K14-positive cells was determined as a percentage of K14-positive cells, as described in the legend to Figure 3 . (A) P90 mice, the percentage of cells coexpressing K12 and K14 among K14 cells increased as the injured corneas healed and reached a plateau of 70% 4 days after debridement, similar to the increase in adult P180 mice. (B) P30 mice, similar results were obtained. The percentage of K12- and K14-positive cells was approximately 70% of all K14-positive cells 7 days after injury, whereas the uninjured contralateral control corneas remained at 30% positive.

Figure 5.

 

X-gal and AP staining of Krt12 ^Cre/Cre /ZAP bitransgenic mice. Corneas from bitransgenic mice at different ages were subjected to histochemical staining for X-gal and alkaline phosphatase activities at (A) P15, (B) P30, (C) P60, (D) P90, (E) P180, and (F) P300. Cells expressing K12 and AP were stained red, whereas the K1-negative, LacZ-positive cells stained blue. At P15, the expression of LacZ (blue) and AP (red) showed a mosaic pattern. At P90 almost all central cornea epithelium expressed K12 and stained red, with sporadic blue cells. At P180 and P300, central corneas were stained red- with blue-stained cells located at the limbus.

Figure 6.

 

P30 corneas after epithelium debridement. Debrided corneas of P30 mice, healed for 24 hours and 7 days, were stained for LacZ and AP, as described in the legend to Figure 5 . Twenty-four hours after debridement (A), Lac Z-positive cells showed a mosaic pattern similar to that of uninjured contralateral cornea (B), and fewer LacZ-positive cells were present in the limbus of the injured corneas. Seven days after injury (C) fewer LacZ-positive cells were found in the limbus and central corneas in comparison to uninjured naïve cornea (D).